School of Distance Education School of Distance Education
Module 1
INTRODUCTION
The term statistics seems to have been derived from the Latin word
‘status’ or Italian word ‘statista’ or the German word ‘statistic, each of
which means political state.
The word ‘Statistics’ is usually interpreted in two ways. The first sense in
which the word is used is a plural noun just refer to a collection of numerical
facts. The second is as a singular noun to denote the methods
generally adopted in the collection and analysis of numerical facts. In the
singular sense the term ‘Statistics’ is better described as statistical methods.
Different authors have defined statistics in different ways. According to
Croxton and Cowden statistics may be defined as ‘‘collection, organisation
presentation, analysis and interpretation of numerical data’’
Population and sample
Population
An aggregate of individual items relating to a phenomenon under
investigation is technically termed as ‘population’. In other words a
collection of objects pertaining to a phenomenon of statistical enquiry is
referred to as population or universe. Suppose we want to collect data
regarding the income of college teachers under University of Calicut,,
then, the totality of these teachers is our population.
In a given population, the individual items are referred to as elementary
units, elements or members of the population. The population has the
statistical characteristic of being finite or infinite. When the number of units
under investigation are determinable, it is called finite population. For
example, the number of college teachers under Calicut University is a finite
population. When the number of units in a phenomenon is indeterminable,
eg, the number of stars in the sky, it is called an infinite population.
Fundamentals of Statistics 4 Fundamentals of Statistics 5
School of Distance Education School of Distance Education
Variables and Attributes
Sample
A quantity which varies from one person to another or one time to
When few items are selected for statistical enquiry, from a given
another or one place to another is called a variable. It is actually a
population it is called a ‘sample’. A sample is the small part or subset of the
numerical value possessed by an item. For example, price of a given
population. Say, for instance, there may be 3000 workers in a factory.
commodity, wages of workers, production and weights of students etc.
One wants to study their consumption pattern. By selecting only 300
workers from the group of 3000, sample for the study has been taken. Attribute means a qualitative characteristic possessed by each individual
This sample is not studied just for its own sake. The motive is to know in a group. It can’t assume numerical values. For example, sex, honesty,
the true state of the population. From the sample study statistical colour etc.
inference about the population can be done. This means that a variable will always be a quantitative characteristic.
Census and sample Method Data concerned with a quantitative variable is called quantitative data and the
data corresponding to a qualitative variable is called qualitative data.
In any statistical investigation, one is interested in studying the
population characteristics. This can be done either by studying the entire We can divide quantitative variables into two (i) discrete (ii) continuous.
items in the population or on a part drawn from it. If we are studying each Those variables which can assume only distinct or particular values are called
and every element of the population, the process is called census method discrete or discontinuous variables. For example, the number of children per
and if we are studying only a sample, the process is called sample survey, family, number rooms in a house etc. Those variables which can take any
sample method or sampling. For example, the Indian population census or numerical value with in a certain range are known as continuous variables.
a socio economic survey of a whole village by a college planning forum Height of a boy is a continuous variable, for it changes continuously in a
are examples of census studies. The national sample survey enquiries are given range of heights of the boys. Similar is the case of weight,: production,
examples of sample studies. price, demand, income, marks etc.
Advantages of Sampling
The sample method is comparatively more economical. Types of Frequency Distribution
The sample method ensures completeness and a high degree of Erricker states “frequency distribution is a classification according to
accuracy due to the small area of operation the number possessing the same values of the variables’’. It is simply a
It is possible to obtain more detailed information, in a sample survey table in which data are grouped into classes and the number of cases
than complete enumeration. which fall in each class is recorded. Here the numbers are usually termed as
Sampling is also advocated where census is neither necessary nor ‘frequencies’. There are discrete frequency distributions and continuous
desirable. frequency distributions.
In some cases sampling is the only feasible method. For example,
1. Discrete Frequency Distribution
we have to test the sharpness of blades-if we test each blade,
perhaps the whole of the product will be wasted; in such If we have a large number of items in the data it is better to prepare a
circumstances the census method will not be suitable. Under these frequency array and condense the data further. Frequency array is prepared
circumstances sampling techniques will be more useful. by listing once and consecutively all the values occurring in the series and
noting the number of times each such value occurs. This is called discrete
A sample survey is much more scientific than census because in it
frequency distribution or ungrouped frequency distribution.
the extent of the reliability of the results can be known where as this
is not always possible in census.
Fundamentals of Statistics Fundamentals of Statistics 7
6
School of Distance Education School of Distance Education
Illustration: The following data give the number of children per family Concepts of a Frequency Table
in each of 25 families 1, 4, 3, 2, 1, 2, 0, 2, 1, 2, 3, 2, 1, 0, 2, 3, 0, 3, 2, 1, i. Class limits: The observations which constitute a class are called
2, 2, 1, 4, 2. Construct a frequency distribution. class limits. The left hand side observations are called lower limits
and the right hand side observations are called upper limits.
No of children Tally marks No of families
0 lll 3 Working classes: The classes of the form 0-9, 10-19, 20-29,... are
1 lllI l 6 called working classes or nominal classes. They are obtained by the
inclusive method of classification where both the limits of a class are
2 llIl IIII 10 included in the same class itself.
3 lllI 4 Actual classes: If we are leaving either the upper limit or the lower
limit from each class, it is called exclusive method of classification.
4 lI 2
The classes so obtained are called ‘actual classes’ or ‘true classes’.
Total 25 The classes 0.5 - 9.5, 9.5 - 19.5, 19.5 - 29.5,... are the actual classes of
the above working classes. The classes of the type 0-10, 10
2. Continuous Frequency Distribution 20, 20 - 30,... are also treated as actual classes. There will be no
break in the actual classes. We can convert working classes to the
An important method of condensing and presenting data is that of the corresponding actual classes using the following steps.
construction of a continuous frequency distribution or grouped frequency
distribution. Here the data are classified according to class intervals. Note the difference between one upper limit and the next lower
limit.
The following are the rules generally adopted in forming a
frequency table for a set of observations. Divide the difference by 2.
Note the difference between the largest and smallest value in the Subtract that value from the lower limits and add the same to the
given set of observations upper limits.
Determine the number classes into which the difference can be For example
divided.
The classes should be mutually exclusive. That means they do not Working Classes Frequency Actual Classes
overlap.
1-2.9 2 0.95-2.95
Arrange a paper with 3 columns, classes, tally marks and frequency.
Write down the classes in the first column. 3-4.9 8 2.95-4.95
5-6.9 10 4.95-6.95
Go though the observations and put tally marks in the respective
classes. 7-8.9 5 6.95-8.95
Write the sum of the tally marks of each class in the frequency Class boundaries: The class limits of the actual classes are called
column. actual class limits or class boundaries.
Note that the sum of the frequencies of all classes should be equal Class mark: The class marks or mid value of classes is the average
to the total number of observations.
Fundamentals of Statistics 9
Fundamentals of Statistics 8
School of Distance Education School of Distance Education
of the upper limit and lower limit of that class. The mid value of Cumulative frequencies are determined on either a less than basis or
working classes and the corresponding actual classes are the same. For more than basis. Thus we get less than cumulative frequencies (<CF) and
example, the class mark of the classes 0 - 9, 10 - 19, 20 - 29,... are greater than or more than cumulative frequencies (>CF). Less than CF
respectively 4.5, 14.5, 24.5,... give the number of observations falling below the upper limit of a class
and greater than CF give the number of observations lying above the
Class interval: The class interval or width of a class is the difference lower limit of the class. Less than CF are obtained by adding successively
between upper limit and lower limit of an actual class. It is better to the frequencies of all the previous classes including the class against
note that the difference between the class limits of a working class is which it is written. The cumulation is started from the lowest size of the
not the class interval. The class interval is usually denoted by ‘c’ or i class to the highest size, (usually from top to bottom). They are based on
or ‘h’. the upper limit of actual classes.
Example More than CF distribution is obtained by finding the cumulation or
total of frequencies starting from the highest size of the class to the lowest
Construct a frequency distribution for the following data
class, (ie., from bottom to top) More than CF are based on the lower limit
70 45 33 64 50 25 65 74 30 20 of the actual classes.
55 60 65 58 52 36 45 42 35 40
51 47 39 61 53 59 49 41 20 55 Classes f UL <CF LL >CF
46 48 52 64 48 45 65 78 53 42
0-10 2 10 2 2 0 3 + 7 + 1 0 + 8 + 5 + 1 23 5
Solution
10-20 5 20 2+5 7 10 3+7+10+8+5 33
Classes Tally marks Frequency 20-30 8 30 2+5+8 15 20 3+7+10+8 28
30-40 10 40 2+5+8+10 25 30 3+7+10 20
20-29 lll 3 40-50 7 50 2+5+8+10+7 32 40 3+10 13
30-39 llll 5 50-60 3 60 2+5+8+10+7+3 35 50 3 3
40-49 llll llll ll 12
50-59 llll llll 10
llll ll
EXERCISES
60-69 7
Multiple Choice Questions
70-79 lll 3
A qualitative characterisic is also known as
Total 40 a. attribute b. variable
Cumulative Frequency Distribution c . variate d. frequency
An ordinary frequency distribution show the number of A variable which assumes only integral values is called
observations falling in each class. But there are instances where we a. continuous b. discrete
want to know how many observations are lying below or above a
particular value or in between two specified values. Such type of c . random d. None of these
information is found in cumulative frequency distributions.
Fundamentals of Statistics 10 Fundamentals of Statistics 11